Preserving non-minimum phase dynamics in model order reduction of fifth-order DC-DC boost converters

Abstract

In this work, a unified modelling approach is developed for the model order reduction of non-minimum phase systems. An optimized approach is adopted to address the problem. The coordinated hunting behavior of Cuban boa snake is made use of to develop a new optimization strategy. A constrained optimization method is developed to reduce a 5th order boost converter in the unified domain. Comparison is carried out with multiple classical techniques as well as some of the widely known nature inspired algorithms. The step and Bode responses using the proposed method offers closeness to the original responses as compared to the existing techniques. The pole zero mapping reveals the non-minimum nature of the reduced system. The stability of the reduced system is reflected through the Nyquist plot. A second-order proportional-integral-derivative (PID) controller is also synthesized using approximate model matching and Cuban boa snake optimization algorithm (CBSOA), which demonstrates superior transient performance, minimal steady-state error, and enhanced robustness.